Although HIV, the causative agent of AIDS, establishes a lifelong infection that cannot be eradicated even with effective treatment, the host immune system has the ability to contain its replication for many years in which the disease remains asymptomatic. The unexpected encounter, in 1995, between the fields of HIV and chemokines has dramatically advanced our understanding of AIDS pathogenesis, opening new perspectives for the development of effective prophylactic and therapeutic measures against HIV. The first HIV-suppressive chemokines, CCL5/RANTES, CCL3/MIP-1&#945; and CCL4/MIP-1&#946;, were identified as major components of the soluble antiviral activity produced by CD8+ T cells and were shown to act via blockade of the principal HIV coreceptor, CCR5. However, subsequent evidence has suggested the existence of other, still unrecognized, HIV-suppressive factors produced by CD8+ T cells and other cells that are activated in the course of immunologic and inflammatory responses. Owing to their inherent antiviral properties, analogues or functional mimics of antiviral chemokines are currently under development as anti-HIV therapeutics or microbicides. 1) Identification and characterization of CXCL4 as a novel HIV-suppressive chemokine. CXCL4/PF4 is an atypical CXC chemokine primarily produced by megakaryocytes and platelets. It is the most abundant protein contained in the platelet &#945;-granules from which it is promptly released upon platelet activation and displays a very high binding affinity for heparin and other glycosaminoglycans (GAGs). We discovered that CXCL4 is a broad-spectrum inhibitor of HIV-1 infection. Treatment with CXCL4 causes dose-dependent inhibition of HIV infection in different experimental models, but unlike the classic HIV-suppressive chemokines, which act by blocking specific coreceptors, CXCL4 is effective on a broad spectrum of HIV-1 isolates, regardless of their coreceptor specificity. We found that CXCL4 inhibits viral attachment and entry through an unconventional mechanism mediated by direct binding to the major viral envelope glycoprotein, gp120. Using a panel of anti-gp120 antibodies of defined specificity, we mapped the CXCL4-binding site to a region in the outer domain of gp120 that is adjacent to, but does not overlap the CD4-binding domain. We also tested a non-allelic CXCL4 variant, CXCL4-L1, whose sequence is nearly identical to that of CXCL4 but has an impaired ability to bind to GAGs, showing that this variant is ineffective against HIV. Thus, even though the antiviral activity of CXCL4 is mediated by direct interaction with the viral envelope and not with the cellular membrane, it seems to be linked to the GAG-binding capacity of this chemokine. 2) Identification and characterization of XCL1 as a novel HIV-suppressive chemokine. Using a large cytokine array screening platform for soluble biomolecules, we identified XCL1/lymphotactin as one of the most expressed chemokines in activated CD8+ T cells. XCL1 is a member of the C (or &#947;) chemokine family, which lack two of the four cysteines that stabilize the classic chemokine fold; hence, XCL1 is a metamorphic protein that can interconvert in solution between two alternative conformations: a classic chemokine fold, which binds and signals through the XCL1 receptor, XCR1, and an alternative fold (all-&#946;), which binds GAGs with high affinity but not XCR1. We found that XCL1 is a conformation-dependent, broad-spectrum inhibitor of HIV-1 that shows striking similarities with the biological features and antiviral mechanism of CXCL4. Analogous to CXCL4, XCL1 inhibits a broad range of HIV-1 isolates in different experimental systems, irrespective of their coreceptor-usage phenotype and genetic subtype. Furthermore, XCL1 blocks HIV-1 through a similar mechanism to that of CXCL4, mediated by direct interaction with the gp120 envelope glycoprotein. However, no specific gp120 region has hitherto been identified as critical for XCL1 binding. Of note, the antiviral activity of XCL1 is selectively associated with the alternatively-folded conformation, emphasizing again, as in the case of CXCL4, the association between HIV blockade and GAG-binding capacity. The structural and functional characterization of XCL1 and its binding site in gp120 may have relevance for the treatment and prevention of HIV-1 infection. 3) Structure-function studies on XCL1. To better define the mode of antiviral activity of XCL1, we conducted a detailed structure-function analysis. Since the HIV-inhibitory activity is associated with the alternative, GAG-binding conformation of XCL1, we mutagenized all the positively charged residues present on the surface of the chemokine. Alanine substitution of either of two residues, Lys42 or Arg43, completely abrogated the antiviral activity of XCL1, as well as its ability to capture HIV-1 virions. Three additional residues, Arg18, Arg35 and Lys46, also play a role, albeit less prominent. Of note, analysis of the crystal structure of XCL1 showed that 4 of the 5 critical residues for anti-HIV activity are clustered in a large positively-charged patch on the surface of the chemokine, which is disrupted in the classic chemokine fold. These residues partially overlap, but are not identical to those previously identified as responsible for the GAG-binding activity of XCL1. Thus, HIV inhibition and GAG binding appear to be related but distinct activities of XCL1. The identification of the structural determinants of antiviral activity of XCL1 may provide a basis for the design of peptide-based or small-molecule inhibitors mimicking this functional region of the chemokine.